Statin-Associated Muscle Symptoms

The VerdictMODERATE CONVICTION

Most statin muscle aches are not caused by the statin.

Work out one number before you do anything else: how long had you been taking the statin when the aching started? In 19 double-blind trials covering 123,940 people, the entire excess of muscle symptoms on statins sat in the first year. After year one there was no excess over a dummy pill at all. If your aching started in year three of a stable dose, the drug is probably not the reason, and something else is going unexamined. If it started in the first year, that is the window worth taking to your prescriber. Separately and urgently: if your urine has turned dark like tea or cola, or you are getting genuinely weaker rather than just sore, that is a same-day medical problem, not a physical therapy one.

Here is what is really happening: across 123,940 people in blinded trials, 27.1 in every 100 on a statin reported muscle pain or weakness, and 26.6 in every 100 on a dummy pill did too.
What most people get wrong: feeling better after you stop the tablet is not proof it was the tablet. In the trial that tested exactly this, relief on stopping was identical whether people had stopped a real statin or a fake one.
Start here: ask your prescriber about a blinded trial, where you take the real tablet and a dummy in random order without knowing which is which. It has been done 200 times in ordinary UK general practice.

Imagine a smoke alarm that goes off both when there is a fire and when someone opens a window. Over a whole building you would find that most of the alarms were windows, and you would still be wrong to rip out the alarm, because some of them were fires. Statin muscle pain works the same way. Across whole populations the drug explains almost none of it, and in the specific people who react and re-react, it explains a real share. The only way to tell which one you are is to close the window without telling the alarm.

Before anything else, work out one number: how long had you been on the statin when the aching started?

In 19 double-blind trials covering 123,940 people, the entire excess of muscle symptoms sat in the first year. After year one there was no excess over a dummy pill at all. Aching that began in year three of a stable dose is probably not the drug, and something else is going unexamined. Separately and urgently: if your urine has turned dark like tea or cola, or you are getting genuinely weaker rather than just sore, that is a same-day medical problem. Go to A&E, not to a physical therapist.

Takes 30 seconds. No equipment needed.

The Verdict

Most statin muscle aches are not caused by the statin. Yours might be. There is a test.

Imagine a smoke alarm that goes off both when there is a fire and when someone opens a window. Survey a whole building and you would find that most of the alarms were windows. You would still be wrong to rip out the alarm, because some of them were fires. Statin muscle pain behaves the same way: across whole populations the drug explains almost none of it, and in the specific people who react and re-react, it explains a real share. The only way to tell which one you are is to close the window without telling the alarm.

Here is what is really happening: across 123,940 people in blinded trials, 27.1 in every 100 taking a statin reported muscle pain or weakness, and 26.6 in every 100 taking a dummy pill did too.
What most people get wrong: feeling better after you stop the tablet is not proof it was the tablet, because in the trial that tested exactly this, relief on stopping was the same whether the tablet had been real or fake.
Start here: ask your prescriber about a blinded trial, where you take the real tablet and a dummy in random order without knowing which is which, because it has already been done 200 times in ordinary UK general practice.

Best for

Anyone on a statin with muscle aching who is thinking about stopping, and anyone whose symptoms started within the first year of treatment.

Skip if

Your urine is dark, you are getting objectively weaker rather than sore, or symptoms have not settled even though you already stopped the statin. Those need a doctor now, not a webpage.

Want the full evidence? Keep scrolling

What Works

Ranked by evidence quality. Note what the top of this list is: not a treatment for the muscle, but a decision about the drug.

Cinematic anatomical study of skeletal muscle tissue

Tier 1 · Strong evidence

Get the attribution question to the prescriber, and name blinded rechallenge. HIGH StatinWISE ran 200 randomised placebo-controlled n-of-1 trials across 50 UK general practices, and two thirds of those who completed intended to restart their statin (Herrett 2021). In SAMSON, half the participants were back on statins six months later, every one of whom had already abandoned them before enrolling (Howard 2021). This is the only intervention in the whole condition that moves the outcome that matters.
Carry on training. HIGH At atorvastatin 80 mg daily for six months, 420 healthy adults showed no change in any measure of muscle strength or exercise capacity (Parker 2013). Across 16 pooled randomised trials there was no effect on maximal heart rate (mean difference 2.8 bpm, 95% CI −7.4 to 13.0) or exercise time (Gadowski 2019), and across 20 studies in adults 65 and over, no relationship with falls or physical activity (Densham 2024). Note precisely what this supports: not detraining someone. It does not make any programme a treatment.
An accurate explanation that is neither dismissal nor confirmation. HIGH Both halves are separately established: most reports are not drug-caused, and a genuine responder minority exists among the people who actually present. Getting that sentence right is the intervention.

Tier 2 and Tier 3 — moderate and emerging

Tier 2 · Moderate evidence

Review the whole medication list at every visit, naming colchicine and antibiotics. MODERATE From two dedicated pharmacovigilance reviews rather than trials, but the severity of the outcome justifies the routine.
Record an objective strength baseline before any withdrawal trial. MODERATE After two months off statins, leg force, endurance and power improved 7.2% to 13.3% across a whole cohort and 8.8% to 16.6% in the symptomatic group, while the symptom score fell from 5.09 to 1.85 (Peyrel 2023). It is the only route to objective evidence in this condition.

Tier 3 · Emerging and thin

Ramp unaccustomed downhill walking and hiking rather than starting abruptly. LOW A single 59-participant trial from 1997: creatine kinase was 62% and 77% higher at 24 and 48 hours after 45 minutes of downhill treadmill walking on lovastatin, with no difference at all after biceps curls in the same subjects (Thompson 1997). The mixed result inside that one trial is the useful part. It is a reason to ramp hill walking, not a reason to avoid resistance training.

What does not work

CoQ10. The default suggestion, and three meta-analyses over an overlapping trial pool disagree with each other. The tie-breaker is population: in the only trial confined to people whose symptoms had actually been confirmed under blinding, pain rose with the statin regardless of CoQ10, despite blood levels rising four-fold (Taylor 2015).
Vitamin D. Two randomised trials null. The positive evidence is entirely non-randomised, stated by the reviewing authors themselves.
Switching from a lipophilic to a hydrophilic statin. No appreciable effect across 135 randomised trials (Irwin 2018), and it remains one of the most widely practised adjustments in this condition.
Dose reduction as a fix for the aching specifically. It genuinely reduces marked enzyme elevation and discontinuation. Its effect on myalgia is a relative risk of 1.04 with a number needed to harm of 173 (Davis 2021). Promising someone that halving the dose will halve the aching is reading the wrong row of the table.
Telling the patient it is in their head. No psychological profile separates confirmed from refuted cases: no differences in anxiety, depression, worry, insomnia or type D personality (Peersen 2021).

Exercise Prescription

Read this before the table. No exercise, stretch, supplement or amount of rest has ever been tested as a treatment for statin muscle symptoms. That is not a gap in this page, it is the state of the research: nobody has run that study. What the evidence does show clearly is that these tablets do not weaken you, do not reduce your fitness and do not raise your falls risk, so the doses below exist to keep you strong. They are not a treatment for the aching. Anything your prescriber tells you overrides this page, and nothing here is a reason to stop taking a statin.

Sit-to-stand general practice 3 × 10 · 3× per week · effort in the thighs is fine, stop if pain is sharp From a dining chair, stand up and sit down slowly, without using your hands if you can.

Step-ups general practice 3 × 10 each leg · 3× per week · aching afterwards is expected, sharp pain is not Step up onto the bottom stair leading with one leg, then down. Swap legs each set.

Heel raises general practice 3 × 12 · 3× per week · calf burning is normal Stand facing a worktop with your hands resting on it for balance. Rise onto your toes and lower slowly.

Walking general practice 20–30 minutes · most days · should not leave you aching more the next day Flat ground, at a pace where you could still hold a conversation.

Doses are ordinary general-practice strength doses taken from a training study in adults aged 65 and over, which used 3 sets of 10 at 80% of one-repetition maximum and 2 sets of 30 at 40% (Alturki 2021). One specific caution: if you are planning a hill walk or a hiking holiday, build up over a few weeks rather than doing it all at once. Long downhill walking is the one exercise interaction with statins that has actually been tested.

Red Flags

Statin muscle symptoms are usually harmless. These five are not, and the last one is not an emergency at all.

Dark cinematic study of muscle tissue under stress

Muscle pain with dark urine, or reduced urine output. Rhabdomyolysis. In one drug-interaction case series, weakness appeared in 82%, dark urine in 71% and myalgia in 61%, mean creatine kinase at diagnosis was 43,890 UI/mL, acute kidney injury occurred in more than half, and 22% of cases were fatal (Bataillard 2019). A&E, same day.
Progressive, symmetric, proximal weakness that is genuinely weak rather than pain-limited. Immune-mediated necrotising myopathy: symmetric in 83.33% of 100 pooled cases, mean creatine kinase 6,853 IU/L, anti-HMGCR antibodies positive in 57 of 57 patients tested (Nazir 2017). GP urgently, for creatine kinase and the antibody.
Muscle symptoms that are not resolving even though the statin has already been stopped. Same condition, and this is its presenting feature. Median time from symptom onset to diagnosis is 3 months, with a documented range out to 156 months (Shelly 2022). GP within days.
A new or increased dose of colchicine, or any new antibiotic, alongside a statin. Of 38 reported adverse events on statin plus colchicine, 25 were myopathy and 10 were rhabdomyolysis (Schwier 2022). Fusidic acid cases began a mean of 30 days after the antibiotic started, which is exactly the window in which people blame their training. Prescriber urgently, and do not load in the meantime.
You are about to stop your statin because of the aching. Not a medical emergency, and arguably the highest-value referral in this whole condition. Stopping trades a symptom for a cardiovascular event risk, on an attribution the randomised evidence says is usually wrong. Prescriber, before you stop, not after.

Refer to: A&E for dark urine or reduced urine output. GP urgently for progressive objective weakness, non-resolution after withdrawal, or a new interacting drug. Prescriber for the attribution question itself.

Return to Training

There is no absence to return from in uncomplicated statin muscle symptoms. These criteria apply only after a red-flag presentation has been medically cleared.

Rhabdomyolysis or myopathy excluded or treated, and the treating physician has cleared exercise
Creatine kinase back within the physician's stated acceptable range, using their threshold rather than any number from this page
No dark urine and normal urine output for at least 7 consecutive days
No progressive decline across two strength measurements taken at least 2 weeks apart
Any interacting drug stopped or reviewed by the prescriber
The attribution conversation has happened, and you know not to stop your statin on your own

How Confident Should You Be

Conviction: MODERATE overall, and it splits hard by endpoint.

HIGH: most reported symptoms are not drug-caused; any excess is confined to the first year; a genuine responder minority exists among people who present; symptom timing and intensity do not discriminate between drug and placebo.

MODERATE to HIGH: higher-intensity regimens raise the risk of marked enzyme elevation (relative risk 4.69, 95% CI 2.50 to 8.80) while barely moving myalgia; statins do not reduce strength, fitness or falls risk.

LOW and unresolved: whether statins blunt aerobic training adaptation. One 37-participant trial says yes, one 26-participant trial says no, and a 16-trial pooled analysis found no effect.

LOW and negative: CoQ10, vitamin D, and SLCO1B1 genotype-guided prescribing.

NO EVIDENCE: any exercise set, repetition, load or frequency parameter for this condition, and any bedside test with published sensitivity or specificity.

What would change my mind about the exercise claim

A randomised trial in 200 to 300 adults whose symptoms have first been confirmed by blinded rechallenge, randomised on continued statin therapy to progressive resistance training, graded aerobic training, or usual care for 12 weeks, with co-primary endpoints of the symptom score and objectively measured leg extensor force, plus statin discontinuation at 12 months. No rehabilitation trial has ever used a confirmed-symptom entry gate, which is exactly why this page carries no exercise prescription for the condition itself.

What would change my mind about the training-adaptation claim

A 200-participant, 12-week randomised trial of aerobic training with and without simvastatin 40 mg in statin-naive adults with metabolic syndrome, powered on change in maximal oxygen uptake, with muscle biopsy citrate synthase as a mechanistic secondary. Both existing trials are under 40 participants, and the widely repeated claim that statins block your training gains rests entirely on one of them.

The Full Picture — Anatomy, Diagnosis & Evidence

What's Actually Going On

Cinematic anatomical rendering of skeletal muscle fibres

Statins block an enzyme called HMG-CoA reductase, which sits at the top of a chemical assembly line. Cholesterol is what that line is famous for making, but it is not the only product. Coenzyme Q10 comes off the same line, and CoQ10 is part of how muscle cells produce energy. Depleting it has been the leading explanation for statin muscle pain for two decades.

The explanation is more confident than the measurements are. When muscle mitochondrial function was measured directly in people whose symptoms had been confirmed by blinded crossover, it looked identical to people whose symptoms had been refuted, at −2.4% against −2.4% (Mangone 2024). And topping the deficiency back up does not fix the symptom: in the only trial restricted to blinded-confirmed cases, raising blood CoQ10 from 1.3 to 5.2 mcg/mL left pain rising with simvastatin exactly as it had before (Taylor 2015).

There is a genuinely separate disease at the far end, and it is not the severe version of the first one. Anti-HMGCR immune-mediated necrotising myopathy is autoimmune: the body makes antibodies against the drug's own target enzyme. It is weak rather than sore, it does not stop when the statin stops, and it needs immune-suppressing treatment. It runs at about 8.3 cases per million person-years (Shelly 2022).

The third mechanism is not pharmacological at all, and by the numbers it is the biggest. In ASCOT-LLA, muscle-related side effects appeared in excess during the unblinded phase of the trial and not during the blinded phase, in the same trial and the same patients (Gupta 2017). That is a real symptom produced by a non-pharmacological route. It is not the same thing as an imaginary one, and the distinction is the whole clinical skill here.

How to Identify It

Dark clinical study of the thigh musculature

There is no bedside test, and the absence is the finding rather than a gap in this page.

Any clinical examination test for drug-caused symptoms Sn: data unavailable · Sp: data unavailable — no such test with published accuracy exists anywhere in 145 retrieved papers
Near-infrared spectroscopy of muscle mitochondrial capacity Sn: data unavailable · Sp: data unavailable — studied for exactly this purpose in a double-blind placebo-controlled crossover and could not tell confirmed from unconfirmed cases (Mangone 2024)
Blinded n-of-1 or crossover rechallenge reference standard by default — prescriber-led, and the only thing that answers the question

Figures do exist for general musculoskeletal clinical prediction rules and they were deliberately not borrowed. Those were not developed in this condition, and a number imported from a different one is an invented number.

What is left is a pattern and a clock. Bilateral, symmetric, lower-limb symptoms appeared in 75% of blinded-confirmed cases against 41% of refuted ones (Peersen 2021). Onset in the first year is where the entire randomised excess sits. Strength is expected to be normal, and creatine kinase normal or trivially raised: across all trials the median shift was about 0.02 times the upper limit of normal, and at the maximum licensed atorvastatin dose for six months, no individual in 420 exceeded ten times normal (Parker 2013).

The Debate

Does the timing story prove the statin caused it?

Conventional practice

Symptoms that start when you begin a statin and ease when you stop it constitute evidence that the statin caused them. This is the basis of everyday stop-start attribution.

SAMSON, Howard 2021, N=60

Sixty people who had already abandoned statins received twelve monthly bottles: four atorvastatin, four placebo, four empty. Scores were 8.0 in the empty months, 16.3 on statin, 15.4 on placebo. Neither symptom intensity on starting a bottle (odds ratio 1.02) nor relief on stopping one (odds ratio 1.01) distinguished real from fake.

Follow the trial. The timing story has no diagnostic power, which retires unblinded stop-start testing as a diagnostic procedure. It does not follow that the patient is imagining it: among people already selected by prior symptoms, intolerance really is more frequent on the drug, at a relative risk of 1.40 with a number needed to harm of 10 (Kraut 2023), and no psychological profile separates confirmed from refuted cases.

Do statins blunt your training gains?

Mikus 2013, N=37

Twelve weeks of aerobic training raised fitness 10% on its own, and only 1.5% when simvastatin 40 mg was added. Muscle citrate synthase rose 13% with training alone and fell in the statin arm.

Kuhlman 2022, N=26; Gadowski 2019, 16 RCTs

After eight weeks of training, maximal oxygen uptake, maximal workload and fat oxidation all rose in the simvastatin arms as well as placebo. A 16-trial pooled analysis found no effect on maximal heart rate or exercise time at all.

Unresolved, and both trials are under 40 participants. The claim that statins block your gains, which circulates widely, rests entirely on the first of the three.

Honest Limitations

The trials answer a population question and the clinic asks an individual one

More than 90% of muscle symptom reports on statins are not caused by the statin, and that fact cannot be applied to any single person, because nobody can tell which report they are looking at. The two designs that can tell them both work, and neither is available in a physical therapy room.

The nocebo evidence gets converted into an instruction it does not support

It licenses distrusting the timing story. It does not license dismissal. Two sentences that sound alike are not alike: "your symptom is real and we do not yet know its cause" is supported by the evidence, and "your symptom is in your head" is not.

A single strength measurement reads the wrong signal

Statins do not reduce measured strength at population level even at the highest licensed dose. Yet objective leg force, endurance and power improved 7.2% to 13.3% across a supervised two-month withdrawal in symptomatic patients. The signal is a change within one person, and obtaining it takes two months and the prescriber's cooperation.

The Nuance

Cinematic anatomical study distinguishing muscle pathologies

The number that should change behaviour here is not about muscles at all. It is that half of the SAMSON participants, every one of whom had already given up on statins before they enrolled, were back on treatment six months after seeing their own blinded data, and two thirds of StatinWISE completers intended to restart.

That reframes the condition. The outcome carrying the consequence is not the ache, it is whether someone is still taking a cardiovascular drug in twelve months. So the harm runs in two directions, and the commoner direction is the quiet one: not a missed myopathy, but an unnecessary discontinuation, traded away for a symptom on an attribution the randomised evidence says is usually wrong.

One more thing the simple answer misses. A study of 390 hypertensive outpatients asked a deceptively simple question: have you ever taken a drug you think gave you muscle symptoms? Nearly half said yes, and the rate was no different between the 250 people who had taken a statin and the 140 who never had. Only age and the total number of drugs predicted it (Sarzani 2024). Muscle aching in this age group is close to a background condition, and a statin is simply the most memorable thing to pin it on.

There is no surgical or procedural pathway. The equivalent decision belongs to the prescriber: continue, rechallenge under blinding, reduce intensity, or move to a non-statin agent. The systematic evidence for that choice was described as lacking until 2025, when 23 studies including 13 randomised trials with 1,868 participants were finally pooled at moderate GRADE confidence (Aebi 2025).

Sources

Nine of the 40 sources behind this page. Full reference list in the clinical protocol card.

Cholesterol Treatment Trialists' Collaboration (2022). Effect of statin therapy on muscle symptoms: an individual participant data meta-analysis. Lancet. PMID 36049498. Individual participant data from 19 double-blind trials, 123,940 participants. Muscle pain or weakness 27.1% on statin against 26.6% on placebo, rate ratio 1.03 (95% CI 1.01 to 1.06). The largest evidence base on this question in existence.
Herrett E, Williamson E, Brack K, et al. (2021). Statin treatment and muscle symptoms: series of randomised, placebo controlled n-of-1 trials. BMJ. PMID 33627334. StatinWISE. 200 randomised n-of-1 trials in UK primary care, 151 analysed. Mean difference statin minus placebo −0.11 (95% CI −0.36 to 0.14).
Howard JP, Wood FA, Finegold JA, et al. (2021). Side Effect Patterns in a Crossover Trial of Statin, Placebo, and No Treatment. J Am Coll Cardiol. PMID 34531021. SAMSON. The only design carrying a no-treatment arm, which is what separates drug effect from expectation from background noise.
Kraut R, Wierenga F, Molstad E, et al. (2023). Intolerance upon statin rechallenge: a systematic review and meta-analysis. PLoS One. PMID 38128013. 8 randomised trials. Intolerance more common on statin, relative risk 1.40 (95% CI 1.23 to 1.60), number needed to harm 10, while the symptom scores did not differ. The disagreement between its two outcomes is the most useful result in this literature.
Parker BA, Capizzi JA, Grimaldi AS, et al. (2013). Effect of statins on skeletal muscle function. Circulation. PMID 23183941. STOMP. 420 healthy statin-naive adults, atorvastatin 80 mg for 6 months. No change in strength or exercise capacity; no individual creatine kinase above ten times normal.
Peyrel P, Mauriège P, Frenette J, et al. (2023). Impact of statin withdrawal on perceived and objective muscle function. PLoS One. PMID 37315062. 92 patients, 2-month withdrawal. Objective leg force, endurance and power improved 7.2% to 13.3% cohort-wide and 8.8% to 16.6% in symptom reporters.
Nazir S, Lohani S, Tachamo N, et al. (2017). Statin-Associated Autoimmune Myopathy: A Systematic Review of 100 Cases. J Clin Rheumatol. PMID 28277343. Symmetric proximal weakness in 83.33%, mean creatine kinase 6,853 IU/L, anti-HMGCR antibody positive in 57 of 57 tested.
Bataillard M, Beyens MN, Mounier G, et al. (2019). Muscle Damage Due to Fusidic Acid-Statin Interaction. Am J Ther. PMID 29189310. 75 cases. Onset a mean of 30 days after the antibiotic started; mean creatine kinase 43,890 UI/mL; fatal in 22%.
Sarzani R, Allevi M, Giulietti F, et al. (2024). Patient-reported muscle symptoms in a hypertensive population eligible for statin therapy. NMCD. PMID 37798232. 390 outpatients. 48.5% reported drug-attributed muscle symptoms, with no difference between those who had ever taken a statin and those who never had.